Complete Generative AI
A personal collection of more than 20 hands-on generative AI projects covering chatbots, retrieval-augmented generation, model fine-tuning and multi-agent systems, built while following a course.
The repository walks through generative AI techniques with runnable, step-by-step implementations, using frameworks and tools such as LangChain, CrewAI and NVIDIA NIM, models including OpenAI's GPT models, Ollama, CodeLlama and Gemma, and databases such as AstraDB, Neo4j and other vector stores. Projects cover conversational chatbots with multi-turn context management, retrieval-augmented document question-answering and hybrid search, model fine-tuning and domain adaptation, and collaborative multi-agent workflows with task delegation.
Each project lives in its own folder with its own README and Jupyter notebooks, and requires its own dependencies installed from a per-project requirements file plus API keys, such as an OpenAI key and an AstraDB application token, set in a local environment file before running.